Multicultural Education, Training & Advocacy, Inc.*
Issue Brief- September 2009
CHARTER SCHOOLS AND ENGLISH LANGUAGE LEARNERS IN MASSACHUSETTS:
POLICY PUSH WITHOUT THE DATA
Introduction
In recent years Massachusetts has struggled, with limited success, to narrow the school success gap
between English Language Learners (ELL) students and native English speakers. The latter place at or
near the top of all students nationwide on the usual achievement measures. ELL students, however lag
far behind.
Some policymakers have proposed that expansion of charter schools in Massachusetts will be the key to
narrowing the ELL student achievement gap. This META bulletin addresses what is currently known
about ELL students and charter schools in the Commonwealth. There is no empirical basis to support
the notion that charters are the cure. To the contrary we find:

English language learner students are conspicuously not represented in the enrollment of
Massachusetts’ charter schools;
 Massachusetts’ charter schools enroll few if any recent immigrant students. Recent
immigrant ELL students need the most help in learning English. These students are
overwhelmingly enrolled in traditional public schools;
 Those ELL students who are enrolled in charter schools are far more likely to have been in
public schools in the United States for 4-5 years or more than are ELL students in traditional
public schools;
 Because of the low numbers of ELL students enrolled, only a very few charter schools have
reported to the Massachusetts Department of Elementary and Secondary Education (DESE)
assessment results on the state’s two main assessments, MCAS and the Massachusetts
English Proficiency Assessment (MEPA);
 The MCAS and MEPA test results for ELL students were reported for only a small fraction of
charter schools and charter school students. They show that in a few instances, the local
charter schools outperform the local school district, in other instances the local school
districts outperform the local charter schools and in other instances there is essentially no
difference.
Other recent research reports that are generally positive about the success of charter schools in
Massachusetts have acknowledged what could be called the charter schools’ ELL deficit.
In sum, there is simply no data to support the claim that in the Commonwealth, more charter schools
will be crucial to addressing the needs of ELL students.
Recommendations
 If for reasons unrelated to the needs of ELL students policymakers decide to significantly
increase the number of charter school openings, it will be essential that the Commonwealth
ensure that: Charter schools demonstrate in their program design that they have programs
in place to address the language and cultural needs of all ELL students and not just those
who are already at the edge of English proficiency;
 Charter schools demonstrate that their administrators and teachers are fully qualified to be
teachers of ELL students including the teaching of English and content instruction to ELL
students;
 Charter schools demonstrate that they have aggressive outreach programs that reach the
parents of potential ELL students in their home language and that assist those parents who
are interested in negotiating the charter school lottery process;
 Charter schools have programs and staffing designed to work with non-English speaking
parents so that they can be full partners in their children’s education.
I.
Charter School Enrollment Does Not Reflect the ELL Student Population
Current state law requires that charter schools not discriminate in membership on the basis of English
language proficiency Ch. 71§ 89(l). Regardless, for whatever reason, ELL students for the most part do
not attend charter schools in numbers reflective of the school districts where the charter schools are
located.
In Boston, for example, approximately 20% of all students are English language learners according to the
MA DESE (18.90%). Yet in 2008-09 Boston Collegiate charter school reported 0% ELL students, as did
Boston Day and Evening, Boston Preparatory, Health Careers Academy, Match Charter, and Smith
Leadership Academy. At eight additional Boston based charter schools the ELL percentages ranged
between 0.7% and 3.8%. Only one Boston charter school, Conservatory Lab charter, approached even
half the district ELL percentage.
This phenomenon is not unique to Boston. The Worcester public schools are 24.3% ELL, the Abby Kelly
Foster charter in Worcester is 3.6% ELL. The Holyoke public schools are 24.2% ELL whereas Holyoke
Community charter school is 10.5% ELL. In Lynn, 25.8% of district students are ELL, at the KIPP charter
META Issue Brief- September 2009 2 | P a g e
school in Lynn, 1.2% of students are reported as ELL. In Somerville, 16.8% of district students are ELL, at
Prospect Hill Academy in Somerville, ELLs count for 1.7% of enrollment. Similar disparities in enrollment
are found in charter schools in Framingham, New Bedford, Haverhill, Lowell, Springfield, Fitchburg,
Everett, and Salem. There are a few exceptions where ELL enrollment approaches or even slightly
exceeds district enrollment, notably in Lawrence and Chelsea.
The bottom line is that in 32 of 40 charter schools located in districts with ELL student populations, the
charter school ELL percentage enrolled does not exceed half of the district ELL enrollment percentage. i
II.
The English Language Learner Enrollment of Charter Schools is Not Reflective of the
General
ELL Student Population
The Massachusetts English Proficiency Assessment (MEPA) is administered to all ELL students and
measures annual growth in acquiring proficiency in English. In Spring 2008, the MA DESE reported that
32,469 ELL students participated in the MEPA statewide-31% of the ELL students had been in school in
the United States for 1 or 2 years. ii MEPA reports results by four levels of English language proficiencyBeginner, Early Intermediate, Intermediate and Transitioning. It is not surprising that ELLs who have
been in this country for the least amount of time are most likely to be classified as Beginners or Early
Intermediates in their English skills. For ELLs who have been in U.S. schools for 1 year, approximately
60% were assessed at the lowest two levels of English proficiency (64% in grades 3-4, 65% in grades 5-6,
66% in grades 7-8 and 58% in grades 9-12). ELLs who had been in U.S. schools for 2 years had greater
English proficiency than more recent immigrant students but significant percentages, between 34% in
grades 3-4 up to 47% in grades 7-8 of second year ELLs were Beginners and Early Intermediates.iii
In contrast only 13% of ELLs in charter schools were reported has having been in U.S. schools for 1 or 2
years. (The 13% figure is skewed by two of the 34 charter schools reporting ELLs, Lowell Community
Charter and the Barnstable Horace Mann School. Only 8% of charter school ELLs in the remaining 32
charter schools have been in U.S. schools for 1 or 2 years.). Consistent with the low percentages of
charter school ELLs who have been in U.S. schools for 1 or 2 years, only 4% of ELLs in charter schools
scored as Beginners on the MEPA compared with 12% of all ELLs assessed statewide. Most (56%) of
ELLs in charter schools have been in U.S. schools for 5 or more years compared with 36% of ELLs
statewide.
These differences can be particularly striking when looked at on a district basis. For example, at
Lawrence Family Day charter school there were 94 reported ELL students in grades 3-8. Of these, 77%
have been in U.S. schools for 5 or more years, 20% have been in U.S. schools for 4 years and 3% (two
students) for 3 years. No students were reported as having been in U.S. schools for 1 or 2 years and no
students were reported as Beginners in their English proficiency.
In the Lawrence public schools at the same grades 42% of ELLs had been in U.S. schools for 1 or 2 years
and only 18% for 5 or more years. Clearly the charter school and the local system in Lawrence are
serving different ELL student populations. What is true in Lawrence is true in other school districts with
large ELL student populations. iv
META Issue Brief- September 2009 3 | P a g e
III.
There is No Obvious Indication that English Language Learner Students in Massachusetts
Charter Schools are Outperforming ELL Students in Local School Districts
Because most ELL students in charter schools tend to have been in the United States for 5 or more years
it would be predicted that their English language skills would result in substantially higher academic
achievement and English language proficiency assessment results than their local school district
counterparts. The fragmentary evidence does not bear out that prediction, at least not in any systemic
sense.
There are two sources of data reported by the Massachusetts Department of Elementary & Secondary
Education, MCAS scores and scores on the Massachusetts English Proficiency Assessment (MEPA). We
do not contend that scores on these two assessments yield a complete picture of how well ELL students
are being educated in either local district or charter schools. That said, the scores are available for
comparison and “they are what they are”.
In 2008 four charter schools enrolled sufficient numbers of ELL students to report MCAS scores in at
least some grades: Holyoke Community Charter School, Lawrence Community Day Charter School,
Lawrence Family Development Charter School and Lowell Community Charter. These results were as
follows: v

English language learner students in the Holyoke public school district outperformed ELLs in
the charter school in Reading in Grade 3 and ELL students in the charter school
outperformed their local district counter parts in Grade 3 Math.
 English language learner students at Lawrence Community Day Charter outperformed their
counterparts in the Lawrence public schools in both Grade 5 Reading/ELA and Grade 5
Math.
 English language learner students at Lawrence Family Development Charter outperformed
their counterparts in the Lawrence public schools in Grades 3, 5, 6 and 7 (both Reading/ELA
and Math). ELL students in the Lawrence Public Schools outperformed their charter school
counterparts in Grade 4 in both subjects.
 English language learner students at Lowell Community Charter did not perform as well as
ELL students in the Lowell Public Schools in Grades 3, 4, 6 or 7 in either Reading or Math and
at grade 5 in ELA. ELLs at the charter school outperformed their local district counterparts
in Grade 5 Math.
In sum, ELL students for the most part did better at the charter schools in Lawrence and worse at the
charter school in Lowell than their local school district counterparts. In Holyoke the picture was mixed.
META Issue Brief- September 2009 4 | P a g e
This ‘mixed bag’ picture is consistent with the results reported in a study commissioned earlier by the
Department of Education. That study, the Charter School Achievement Comparison Study: MCAS 20012005 was limited to six charter schools. The comparisons reported showed that:
 In Boston, English language learner students at the Boston Renaissance charter school did
not score as well in English Language Arts (ELA) as their district counterparts but did
outperform the latter in Math. In one year, 2005, ELL students at the Uphams Corner
charter outperformed district ELL students in Math.
 In Lowell, the local school district ELL students outperformed ELL students at Lowell
Community charter in two of three years reported in Math and in three of four years
reported in ELA.
 In Worcester, ELL students in the local school district outperformed ELLs at Seven Hills
charter school in Math and ELLs at the charter outperformed their district counterparts in
ELA.
 ELL students at both Lawrence charter schools outperformed their counterparts in the
Lawrence public schools.
 In one year, 2005, ELL students at the Banneker charter school in Cambridge outperformed
ELLs in Boston and Cambridge in Math.
In sum, English language learner students in charter schools performed better than district ELL students
in Lawrence and worse than district ELL students in Lowell and results were mixed in Boston and
Worcester.
These results are not presented as clear proof one way or the other. No sophisticated statistical tests
were performed of levels of educational significance. The point, however, is that what we have
presented here constitutes the public record, available at the MA DESE website. What the results do
show is that there is very little data available to demonstrate how well or poorly charter schools in the
Commonwealth are doing to educate ELL students. And what data is available presents a decidedly
mixed picture. That there is a glaring gap in achievement between ELL students and their non-ELL is
undeniable. But if more charter schools is the answer, then based on this data, the question must be
something other than “what can the Commonwealth do to close the ELL achievement gap?”
IV.
Four Criteria for Charters to Ensure Meaningful Access Including Appropriate Programs for
English Language Learner Students
If charter schools are to become more than a marginal and theoretical educational option for ELL
students they must in the future:
1) Demonstrate that in their program design the charter schools have programs in place to
address the language and cultural needs of all ELL students and not just those ELL students
who are already at the edge of English proficiency;
META Issue Brief- September 2009 5 | P a g e
2) Demonstrate that the charter schools’ administrators and teachers are fully qualified to be
teachers of ELL students, this means that teachers of ELL students in charter schools include
both ESL teachers and content teachers who meet state standards for ELL teaching;
3) Demonstrate that charter schools have an aggressive and relevant outreach program that
reaches the parents of potential ELL students in their home language and that assists those
parents who are interested in negotiating the charter school lottery process;
4) Demonstrate that charter schools have programs and staffing designed to work with nonEnglish speaking parents so that they can be full partners in their children’s education.
V.
Current Charter School Policy Proposals, Legislative Bills, and Ballot Initiatives, and
Language
Learner Students in Massachusetts
We look briefly at three prominent proposals to see whether they meet these four criteria. The
proposals are: “An Act Relative to Charter Schools in Underperforming Districts” (the Governor’s
proposal); An act to increase quality and parental choice in public education by expanding enrollment in
high-quality public charter schools” (09-12) and “An act to improve public education by expanding
enrollment in high-quality public charter schools” (09-13)(ballot initiatives); An Act to Reform Education
In Public Schools Through the Creation Of In-District Charter Schools With Pay-For-Performance (Boston
Mayor Thomas Menino’s bill).
The Governor’s proposal makes an attempt to address the needs of ELL students in several ways. Such
students are among the categories of students whose success is to be considered in evaluating charter
school applications. Moreover, applicants must propose a detailed recruitment plan setting forth
specific strategies the school would use to attract and enroll a student population that has a student
population comparable to the same area of the district in which the charter school is to be located.
Charter schools must also file a report with the board of elementary and secondary education detailing
progress in meeting enrollment and retention goals and the report must include an accounting of how
many students were designated as ELL upon enrollment and how many were subsequently no longer so
designated. Finally, the bill provides that there will be a mailing to the parents of all students in the
most prevalent languages of the district informing parents about the charter school process.
While the thrust of these proposals might result in more ELL students attending charter schools than at
present, the Governor’s proposal undercuts that thrust in several ways.
 First, a charter school recruitment plan would have to target no more than 2 of several subcategories: low-income, special education, limited English proficient, all students scoring
less than proficient on the MCAS and at-risk students. In other words, a charter school,
even one situated in a heavily immigrant neighborhood, could be approved without
recruiting a single English language learner student provided that other low-income
students who scored as Needs Improvement on the MCAS were taken.
 Second, while “limited English proficient” is among the possible categories to be recruited,
there is nothing to focus recruitment and retention efforts on those limited English
proficient students with the most limited English. Limited English proficient/ELL students
who have been in U.S. schools for five or more years and who are already at the
META Issue Brief- September 2009 6 | P a g e
Transitioning or highest level of English proficiency, could continue to account for all or
most of a charter school’s ELL student population.
 Third, by making the focus of a charter school’s reporting how many ELL students were no
longer designated as needing English language services, such schools would have a heavy
incentive to limit ELL recruitment to those already at high levels of English proficiency, i.e.
those closest to being exited from the ELL designation.
 Fourth, nothing in the Governor’s proposal requires that charter schools have programs and
staff in place to specifically address the needs of ELL students. Charter schools that have
trained staff, specially designed programs and the means to involve non-English speaking
parents in the educational program will be much more likely to attract those parents than
those that do not demonstrate their interest in serving the ELL student population through
such concrete measures.
There are two other proposals. House Bill No. 4166, An Act to Reform Education In Public Schools
Through the Creation Of In-District Charter Schools With Pay-For-Performance was filed on July 7. The
bill mentions English language learners in one provision, their annual “progress” is among fourteen
areas to be looked at as part of a “performance contract” to be entered into with “eligible applicants”.
There is no further description. Existing federal and state law already require that the annual progress
of ELL students be measured and reported and the reference in H. 4166 would seem to add nothing to
those existing provisions.
Finally, two ballot questions have been filed with the office of the attorney general. Depending upon
whether they are certified by the attorney general and further whether sufficient valid signatures are
gathered by their proponents, one or both of these measures may come before the legislature (and
ultimately the voters). The measures are entitled: “An act to increase quality and parental choice in
public education by expanding enrollment in high-quality public charter schools” (09-12) and “An act to
improve public education by expanding enrollment in high-quality public charter schools” (09-13). Each
proposals states that applicants for a charter will have to demonstrate academic success in serving
significant numbers of students from 1 or more of 7 categories. Students with limited English
proficiency are enumerated as one of the seven categories. There is no definition of “significant
numbers of students” and no definition of what level of English proficient student the applicant had ever
served. There is also no requirement that any English language learner student actually be enrolled in
the charter school since they are only one (1) category among seven (7). None of the other program,
staffing or parental access criteria is addressed in these proposals.
* Multicultural Education, Training & Advocacy (META, Inc.) founded in 1983 is a tax-exempt non-profit,
national advocacy organization specializing in education issues affecting low-income, language minority, and
immigrant youth.
i
See, http://profiles.doe.mass.edu/state_report/selectedpopulations.aspx
http://www.doe.mass.edu/mcas/mepa/results.html, Table 1. See generally the school and district MEPA results
found at http://www.doe.mass.edu/mcas/mepa/results.html for data reported in this section.
iii
At previous endnote, Tables 3-6.
iv
In a recent study of Charters and Pilot schools in Boston prepared for the Boston Foundation, an introductory
caveat states that “English learners may know no English at all or have some proficiency. It is possible that Pilot and
Charter Schools serve different proportions of these subgroups.” Abdulkadiroglu, Angrist, Dynarkski, Fullerton,
ii
META Issue Brief- September 2009 7 | P a g e
Kane and Pathak, Informing the Debate (2009). The authors noted that because of limitations in the data set with
which they were working, they were unable to analyze those language proficiency differences in their study.
v
The Department does not report scores unless there are at least 10 students in a subgroup at a given grade level.
The MA DESE study states that: “there were few charter schools with ...a Limited English Proficient
subgroup…large enough to support comparisons” at p. 10.
META Issue Brief- September 2009 8 | P a g e